Aplastic anemia
- a population-based study of
epidemiology, treatment, and
prognostic factors
Krista VahtDepartment of Internal Medicine and Clinical Nutrition Institute of Medicine
Sahlgrenska Academy, University of Gothenburg
Cover illustration: Bone marrow in aplastic anemia (magnification: ~
40x)
Aplastic anemia – a population-based study of epidemiology, treatment, and prognostic factors
© Krista Vaht 2020 krista.vaht@gu.se
ISBN 978-91-7833-970-9 (PRINT) ISBN 978-91-7833-971-6 (PDF) Printed in Borås, Sweden 2020 Printed by Stema Specialtryck AB
SVANENMÄRKET
Trycksak 3041 0234
Cover illustration: Bone marrow in aplastic anemia (magnification: ~
40x)
Aplastic anemia – a population-based study of epidemiology, treatment, and prognostic factors
© Krista Vaht 2020 krista.vaht@gu.se
ISBN 978-91-7833-970-9 (PRINT) ISBN 978-91-7833-971-6 (PDF) Printed in Borås, Sweden 2020 Printed by Stema Specialtryck AB
epidemiology, treatment, and prognostic factors
Krista Vaht
Department of Internal Medicine and Clinical Nutrition, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Sweden
Abstract
Background and aims. Aplastic anemia (AA) is a rare but life-threatening
disease. The introduction of immunosuppressive treatment (IST) and hematopoietic stem cell transplantation (HSCT) has considerably improved the outcome of patients with AA. However, modern-day population-based data are limited. This thesis aimed to retrospectively analyze the incidence, treatment modalities, survival, and immune markers in the bone marrow of patients diagnosed with AA in Sweden from 2000–2011.
Patients and methods. Patients were included via the National Patient
Registry and diagnosed according to the Camitta criteria. All data were collected from medical charts. In paper IV, immunohistochemistry was used to obtain data on regulatory T cells and macrophages in the bone marrow.
Results and conclusions. We identified 257 confirmed cases, with an overall
incidence of 2.35 cases per million inhabitants per year. The 5-year overall survival (OS) was >90% in patients aged up to 39 years but 38.1% in patients aged ≥60 years. Multivariate analysis showed that age ≥40 years, very severe AA, and no specific therapy were independent risk factors for inferior survival. First-line IST treated patients (n=158) showed a 47% response rate with no difference regarding the age groups or anti-thymocyte globulin (ATG) formulation. The response was significantly associated with the severity grade at the time of treatment initiation, and very severe AA patients exhibited a response rate of 22%. Sixty-eight patients underwent HSCT with a 5-year OS of 86.8%. The graft-versus-host-disease-free, relapse/rejection-free survival at 5 years was 69.1%. Patients aged ³40 years had higher transplant-related mortality that translated into a lower 5-year OS. In paper IV, we found lower numbers of FOXP3-positive regulatory T cells in AA patients without predictive value for IST response and that patients with a higher number of CD163-positive macrophages had a better 5-year OS, but this benefit was only observed in the non-severe AA group. In conclusion, younger patients have very good long-term survival regardless of the choice of therapy, whereas the outcome for patients ≥60 years remains poor. Very severe AA patients respond poorly to ATG, which indicates the need for a different treatment approach.
Keywords: aplastic anemia, real-world data, ATG, HSCT, regulatory T cells ISBN 978-91-7833-970-9 (PRINT) ISBN 978-91-7833-971-6 (PDF)
Aplastisk anemi (AA) är en sällsynt men mycket allvarlig blodsjukdom. Den orsakar låga blodvärden och cellfattig benmärg. AA har i de flesta fall en autoimmun orsak och svarar ofta på immunhämmande behandling (IST) med antithymocytglobulin (ATG). Sjukdomen har olika svårighetsgrader, som beror på nivån av vita blodkroppar i blodet. Man har kunnat påvisa en minskad förekomst av regulatoriska T-celler (Tregs) och ökad förekomst av makrofager hos AA-patienter. Vi har retrospektivt studerat incidens, överlevnad, behandlingsregimer och immunologiska markörer i ett stort befolkningsbaserat, oselekterat patientmaterial med AA som insjuknade i Sverige under åren 2000-2011. Vi observerade att incidensen på AA inte har ändrat sig de senaste 30 åren: i Sverige insjuknar 2,35 personer per miljon invånare och år. Vi identifierade sammanlagt 257 AA patienter med en medianålder på 60 år. Primär behandling i form av IST gavs till 63%, 10% benmärgstransplanterades, och 27% fick palliativ eller ingen behandling alls. Vi såg att 5-års-överlevnaden tydligt påverkades av ålder och att ålder >40 år, mycket svår AA (VSAA) och palliativ behandling/ingen behandling alls var oberoende riskfaktorer för sämre överlevnad. Vidare har vi analyserat behandling med ATG (n=158) och sett att knappt 50% svarar på sådan behandling. Behandlingssvaret var tydligt beroende av svårighetsgraden: patienter med VSAA svarade sämst. Benmärgstransplanterade patienter (n=68) hade en 5-års-överlevnad på 86,8% och överlevnaden påverkades av patienternas ålder: 92% för patienter <40 år och 71% för patienter ³40. Transplantat-mot-mottagare sjukdom (GvHD) kan vara en mycket allvarlig komplikation efter transplantation. Vi analyserade därför det nya kombinerade utfallsmåttet GvHD och avstötning/återfallsfri överlevnad (GRFS) vilket tidigare inte gjorts vid aplastisk anemi och såg en 5-års-GRFS på 69%. Patienter över 40 år hade en sämre GRFS (53%). Till slut analyserade vi immunmarkörer (Tregs och makrofager) i benmärgen på AA patienter och friska kontroller. Vi fann mycket låga Tregs hos AA patienter utan samband med behandlingssvar. Samtidigt såg vi att patienter med högre antal makrofager hade bättre överlevnad, i synnerhet icke-svåra AA patienter. Vi kan konkludera att unga patienter klarar sig bra, oavsett primär behandling och att den största utmaningen idag är förbättra behandlingen för lite äldre patienter och patienter med mycket svår aplastisk anemi.
epidemiology, treatment, and prognostic factors
Krista Vaht
Department of Internal Medicine and Clinical Nutrition, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Sweden
Abstract
Background and aims. Aplastic anemia (AA) is a rare but life-threatening
disease. The introduction of immunosuppressive treatment (IST) and hematopoietic stem cell transplantation (HSCT) has considerably improved the outcome of patients with AA. However, modern-day population-based data are limited. This thesis aimed to retrospectively analyze the incidence, treatment modalities, survival, and immune markers in the bone marrow of patients diagnosed with AA in Sweden from 2000–2011.
Patients and methods. Patients were included via the National Patient
Registry and diagnosed according to the Camitta criteria. All data were collected from medical charts. In paper IV, immunohistochemistry was used to obtain data on regulatory T cells and macrophages in the bone marrow.
Results and conclusions. We identified 257 confirmed cases, with an overall
incidence of 2.35 cases per million inhabitants per year. The 5-year overall survival (OS) was >90% in patients aged up to 39 years but 38.1% in patients aged ≥60 years. Multivariate analysis showed that age ≥40 years, very severe AA, and no specific therapy were independent risk factors for inferior survival. First-line IST treated patients (n=158) showed a 47% response rate with no difference regarding the age groups or anti-thymocyte globulin (ATG) formulation. The response was significantly associated with the severity grade at the time of treatment initiation, and very severe AA patients exhibited a response rate of 22%. Sixty-eight patients underwent HSCT with a 5-year OS of 86.8%. The graft-versus-host-disease-free, relapse/rejection-free survival at 5 years was 69.1%. Patients aged ³40 years had higher transplant-related mortality that translated into a lower 5-year OS. In paper IV, we found lower numbers of FOXP3-positive regulatory T cells in AA patients without predictive value for IST response and that patients with a higher number of CD163-positive macrophages had a better 5-year OS, but this benefit was only observed in the non-severe AA group. In conclusion, younger patients have very good long-term survival regardless of the choice of therapy, whereas the outcome for patients ≥60 years remains poor. Very severe AA patients respond poorly to ATG, which indicates the need for a different treatment approach.
Keywords: aplastic anemia, real-world data, ATG, HSCT, regulatory T cells ISBN 978-91-7833-970-9 (PRINT) ISBN 978-91-7833-971-6 (PDF)
Aplastisk anemi (AA) är en sällsynt men mycket allvarlig blodsjukdom. Den orsakar låga blodvärden och cellfattig benmärg. AA har i de flesta fall en autoimmun orsak och svarar ofta på immunhämmande behandling (IST) med antithymocytglobulin (ATG). Sjukdomen har olika svårighetsgrader, som beror på nivån av vita blodkroppar i blodet. Man har kunnat påvisa en minskad förekomst av regulatoriska T-celler (Tregs) och ökad förekomst av makrofager hos AA-patienter. Vi har retrospektivt studerat incidens, överlevnad, behandlingsregimer och immunologiska markörer i ett stort befolkningsbaserat, oselekterat patientmaterial med AA som insjuknade i Sverige under åren 2000-2011. Vi observerade att incidensen på AA inte har ändrat sig de senaste 30 åren: i Sverige insjuknar 2,35 personer per miljon invånare och år. Vi identifierade sammanlagt 257 AA patienter med en medianålder på 60 år. Primär behandling i form av IST gavs till 63%, 10% benmärgstransplanterades, och 27% fick palliativ eller ingen behandling alls. Vi såg att 5-års-överlevnaden tydligt påverkades av ålder och att ålder >40 år, mycket svår AA (VSAA) och palliativ behandling/ingen behandling alls var oberoende riskfaktorer för sämre överlevnad. Vidare har vi analyserat behandling med ATG (n=158) och sett att knappt 50% svarar på sådan behandling. Behandlingssvaret var tydligt beroende av svårighetsgraden: patienter med VSAA svarade sämst. Benmärgstransplanterade patienter (n=68) hade en 5-års-överlevnad på 86,8% och överlevnaden påverkades av patienternas ålder: 92% för patienter <40 år och 71% för patienter ³40. Transplantat-mot-mottagare sjukdom (GvHD) kan vara en mycket allvarlig komplikation efter transplantation. Vi analyserade därför det nya kombinerade utfallsmåttet GvHD och avstötning/återfallsfri överlevnad (GRFS) vilket tidigare inte gjorts vid aplastisk anemi och såg en 5-års-GRFS på 69%. Patienter över 40 år hade en sämre GRFS (53%). Till slut analyserade vi immunmarkörer (Tregs och makrofager) i benmärgen på AA patienter och friska kontroller. Vi fann mycket låga Tregs hos AA patienter utan samband med behandlingssvar. Samtidigt såg vi att patienter med högre antal makrofager hade bättre överlevnad, i synnerhet icke-svåra AA patienter. Vi kan konkludera att unga patienter klarar sig bra, oavsett primär behandling och att den största utmaningen idag är förbättra behandlingen för lite äldre patienter och patienter med mycket svår aplastisk anemi.
List of papers
This thesis is based on the following studies, referred to in the text by their Roman numerals.
I. Vaht K, Göransson M, Carlson K, Isaksson C, Lenhoff S, Sandstedt A, Uggla B, Winiarski J, Ljungman P, Brune M, Andersson P-O. Incidence and outcome of acquired aplastic anemia: real-world data from patients diagnosed in Sweden from 2000-2011.
Haematologica. 2017;102(10):1683-1690.
II. Vaht K, Göransson M, Carlson K, Isaksson C, Lenhoff S, Sandstedt A, Uggla B, Winiarski J, Ljungman P, Brune M, Andersson P-O. Low response rate to ATG-based immunosuppressive therapy in very severe aplastic anaemia - A Swedish nationwide cohort study. European J Haematology. 2018;100(6):613-620.
III. Vaht K, Göransson M, Carlson K, Isaksson C, Lenhoff S, Sandstedt A, Uggla B, Winiarski J, Ljungman P, Andersson P-O, Brune M. High Graft-versus-Host Disease-Free, Relapse/Rejection-Free Survival and Similar Outcome of Related and Unrelated Allogeneic Stem Cell Transplantation for Aplastic Anemia: A Nationwide Swedish Cohort Study. Biology of Blood and Marrow Transplantation. 2019;25(10):1970-1974. IV. Vaht K, Brenner J, Bram Ednersson S, Ljungman P,
BruneM and Andersson P-O. Bone marrow expression of CD68/CD163 macrophages, IL-17 and FOXP3 cells in aplastic anemia and their relation to prognosis. Manuscript
List of papers
This thesis is based on the following studies, referred to in the text by their Roman numerals.
I. Vaht K, Göransson M, Carlson K, Isaksson C, Lenhoff S, Sandstedt A, Uggla B, Winiarski J, Ljungman P, Brune M, Andersson P-O. Incidence and outcome of acquired aplastic anemia: real-world data from patients diagnosed in Sweden from 2000-2011.
Haematologica. 2017;102(10):1683-1690.
II. Vaht K, Göransson M, Carlson K, Isaksson C, Lenhoff S, Sandstedt A, Uggla B, Winiarski J, Ljungman P, Brune M, Andersson P-O. Low response rate to ATG-based immunosuppressive therapy in very severe aplastic anaemia - A Swedish nationwide cohort study. European J Haematology. 2018;100(6):613-620.
III. Vaht K, Göransson M, Carlson K, Isaksson C, Lenhoff S, Sandstedt A, Uggla B, Winiarski J, Ljungman P, Andersson P-O, Brune M. High Graft-versus-Host Disease-Free, Relapse/Rejection-Free Survival and Similar Outcome of Related and Unrelated Allogeneic Stem Cell Transplantation for Aplastic Anemia: A Nationwide Swedish Cohort Study. Biology of Blood and Marrow Transplantation. 2019;25(10):1970-1974. IV. Vaht K, Brenner J, Bram Ednersson S, Ljungman P,
BruneM and Andersson P-O. Bone marrow expression of CD68/CD163 macrophages, IL-17 and FOXP3 cells in aplastic anemia and their relation to prognosis. Manuscript
Content
ABBREVIATIONS ... III
DEFINITIONS IN SHORT ... V
1 INTRODUCTION ... 1
1.1 Definitions and diagnostic criteria ... 2
1.2 Epidemiology ... 2 1.3 Pathophysiology ... 3 1.3.1 Immunological mechanisms ... 3 1.3.2 Telomeres ... 4 1.3.3 Clonality ... 5 1.4 Differential diagnosis ... 7
1.4.1 Inherited bone marrow failure syndromes ... 7
1.4.2 Hypoplastic myelodysplastic syndrome ... 9
1.5 Treatment ... 11
1.5.1 Hematopoietic stem cell transplantation ... 12
1.5.2 Immunosuppressive treatment ... 15
1.6 Population-based studies ... 18
2 AIMS ... 20
3 PATIENTS AND METHODS ... 21
4 RESULTS ... 24 5 DISCUSSION ... 31 6 CONCLUSION ... 34 7 FUTURE PERSPECTIVES ... 35 ACKNOWLEDGEMENT ... 36 REFERENCES ………....37 Papers I-IV
Abbreviations
AA Aplastic anemiaALC Absolute lymphocyte count AML Acute myeloid leukemia ARC Absolute reticulocyte count ATG Anti-thymocyte globulin
BM Bone marrow
cGvHD Chronic Graft-versus-Host disease
CR Complete remission
CsA Cyclosporine A
DC Dyskeratosis congenita
FA Fanconi anemia
GPI Glycosylphosphatidylinositol
GRFS Graft-versus-host disease free, rejection-free survival GvHD Graft-versus-Host disease
hATG Horse anti-thymocyte globulin HLA Human leukocyte antigen
hMDS Hypoplastic myelodysplastic syndrome HSCT Hematopoietic stem cell transplantation HPF High power field
IAAS The International Agranulocytosis and Aplastic Anemia Study IBMFS Inherited bone marrow failure syndromes IL-17 Interleukin 17
Content
ABBREVIATIONS ... III
DEFINITIONS IN SHORT ... V
1 INTRODUCTION ... 1
1.1 Definitions and diagnostic criteria ... 2
1.2 Epidemiology ... 2 1.3 Pathophysiology ... 3 1.3.1 Immunological mechanisms ... 3 1.3.2 Telomeres ... 4 1.3.3 Clonality ... 5 1.4 Differential diagnosis ... 7
1.4.1 Inherited bone marrow failure syndromes ... 7
1.4.2 Hypoplastic myelodysplastic syndrome ... 9
1.5 Treatment ... 11
1.5.1 Hematopoietic stem cell transplantation ... 12
1.5.2 Immunosuppressive treatment ... 15
1.6 Population-based studies ... 18
2 AIMS ... 20
3 PATIENTS AND METHODS ... 21
4 RESULTS ... 24 5 DISCUSSION ... 31 6 CONCLUSION ... 34 7 FUTURE PERSPECTIVES ... 35 ACKNOWLEDGEMENT ... 36 REFERENCES ………....37 Papers I-IV
Abbreviations
AA Aplastic anemiaALC Absolute lymphocyte count AML Acute myeloid leukemia ARC Absolute reticulocyte count ATG Anti-thymocyte globulin
BM Bone marrow
cGvHD Chronic Graft-versus-Host disease
CR Complete remission
CsA Cyclosporine A
DC Dyskeratosis congenita
FA Fanconi anemia
GPI Glycosylphosphatidylinositol
GRFS Graft-versus-host disease free, rejection-free survival GvHD Graft-versus-Host disease
hATG Horse anti-thymocyte globulin HLA Human leukocyte antigen
hMDS Hypoplastic myelodysplastic syndrome HSCT Hematopoietic stem cell transplantation HPF High power field
IAAS The International Agranulocytosis and Aplastic Anemia Study IBMFS Inherited bone marrow failure syndromes IL-17 Interleukin 17
IST Immunosuppressive treatment MDS Myelodysplastic syndrome
MΦ Macrophage
NSAA Non-severe aplastic anemia OR Overall response
OS Overall survival
PBSC Peripheral blood stem cells
PNH Paroxysmal nocturnal hemoglobinuria PR Partial remission
rATG Rabbit anti-thymocyte globulin
RNA Ribonucleic acid, a biological macromolecule SAA Severe aplastic anemia
TRM Transplant related mortality Tregs Regulatory T cells
TNF-a Tumor necrosis factor-a URD Unrelated donor
VSAA Very severe aplastic anemia
Definitions in short
Severe AA Bone marrow cellularity <25%, or 25%-50% with <30% residual hematopoietic cells and 2/3 of the following: neutrophil count <0.5×109/l; platelet count
<20×109/l; reticulocyte count <20×109/l
(Camitta, 1975)
Very severe AA As for severe AA but neutrophils <0.2×109/l (Bacigalupo 1988).
Non-severe AA Not fulfilling the criteria for severe or very severe aplastic anemia (Camitta, 1975)
Complete remission Hemoglobin normal for age; neutrophil count >1.5×109/l; platelet count
>150×109/l (Camitta, 2000)
Partial remission Transfusion independent, no longer meeting criteria for severe disease (Camitta, 2000)
IST Immunosuppressive treatment MDS Myelodysplastic syndrome
MΦ Macrophage
NSAA Non-severe aplastic anemia OR Overall response
OS Overall survival
PBSC Peripheral blood stem cells
PNH Paroxysmal nocturnal hemoglobinuria PR Partial remission
rATG Rabbit anti-thymocyte globulin
RNA Ribonucleic acid, a biological macromolecule SAA Severe aplastic anemia
TRM Transplant related mortality Tregs Regulatory T cells
TNF-a Tumor necrosis factor-a URD Unrelated donor
VSAA Very severe aplastic anemia
Definitions in short
Severe AA Bone marrow cellularity <25%, or 25%-50% with <30% residual hematopoietic cells and 2/3 of the following: neutrophil count <0.5×109/l; platelet count
<20×109/l; reticulocyte count <20×109/l
(Camitta, 1975)
Very severe AA As for severe AA but neutrophils <0.2×109/l (Bacigalupo 1988).
Non-severe AA Not fulfilling the criteria for severe or very severe aplastic anemia (Camitta, 1975)
Complete remission Hemoglobin normal for age; neutrophil count >1.5×109/l; platelet count
>150×109/l (Camitta, 2000)
Partial remission Transfusion independent, no longer meeting criteria for severe disease (Camitta, 2000)
1 Introduction
Aplastic anemia (AA) is a rare and potentially lethal disease. It is characterized by empty bone marrow and pancytopenia. More than 100 years ago (1888), Paul Ehrlich described the autopsy of a young woman who had a short fatal illness characterized by anemia, bleeding, and infection accompanied by yellow instead of red bone marrow in her
femur1. The disease was first named pernicious to reflect its fatal course.
In 1904, Vaquez and Aubertin described another fatal case in which histological examination revealed fatty bone marrow with few
lymphocytes, and the term AA was used for the first time2. Clinical
disease features were described by Richard C. Cabot and other
pathologists in the early 20th century3. Factors responsible for AA
development have changed over time. At the beginning of the 1900s, the industrial exposure of benzene and subsequent development of AA were reported among Swedish bicycle factory workers and Newark (USA)
leatherworkers4,5. In the 1960s, several AA patients were diagnosed
during the introduction of chloramphenicol,6,7 and the direct toxic effect
was assumed to be responsible for the occurrence of AA. Since the 1970s, after the first bone marrow transplantation and the use of anti-lymphocyte globulin (ATG) to treat AA, evolving evidence for a possible immunological mechanism contributing to AA has been reported. Today, immunological bone marrow destruction is considered the main underlying mechanism in AA.
1 Introduction
Aplastic anemia (AA) is a rare and potentially lethal disease. It is characterized by empty bone marrow and pancytopenia. More than 100 years ago (1888), Paul Ehrlich described the autopsy of a young woman who had a short fatal illness characterized by anemia, bleeding, and infection accompanied by yellow instead of red bone marrow in her
femur1. The disease was first named pernicious to reflect its fatal course.
In 1904, Vaquez and Aubertin described another fatal case in which histological examination revealed fatty bone marrow with few
lymphocytes, and the term AA was used for the first time2. Clinical
disease features were described by Richard C. Cabot and other
pathologists in the early 20th century3. Factors responsible for AA
development have changed over time. At the beginning of the 1900s, the industrial exposure of benzene and subsequent development of AA were reported among Swedish bicycle factory workers and Newark (USA)
leatherworkers4,5. In the 1960s, several AA patients were diagnosed
during the introduction of chloramphenicol,6,7 and the direct toxic effect
was assumed to be responsible for the occurrence of AA. Since the 1970s, after the first bone marrow transplantation and the use of anti-lymphocyte globulin (ATG) to treat AA, evolving evidence for a possible immunological mechanism contributing to AA has been reported. Today, immunological bone marrow destruction is considered the main underlying mechanism in AA.
1.1 Definition and diagnostic criteria
AA is sometimes referred to as idiopathic, acquired, or immune. It is defined as pancytopenia with hypocellular bone marrow in the absence of fibrosis or an abnormal infiltrate. For the diagnosis of AA, at least two of the three following criteria must be present: hemoglobin <100 g/l, platelet count <50×109/l, or neutrophil count <1.5×109/l8,9.
To assess AA severity, modified Camitta criteria are used8,10:
Severe AA (SAA) is defined as bone marrow cellularity <25% or 25%– 50% with <30% residual hematopoietic cells and two of the following three criteria: neutrophil count <0.5×109/l, platelet count <20×109/l, or
reticulocyte count <20×109/l. Very severe AA (VSAA) is similar to
SAA but with a neutrophil count <0.2×109/l. Non-severe AA (NSAA)
does not fulfill the criteria for SAA or VSAA.
1.2 Epidemiology
The incidence of AA varies considerably in different parts of the world and has also changed over the years. Historically, reports from Europe and the United States in 1960–70 showed diverse results; incidence rates of up to 6–10 AA cases per million people per year. In some of these studies, an association with toxic agriculture substances was shown11-13.
However, different diagnostic criteria were used, and some cases represented other diagnoses11. The International Agranulocytosis and
Aplastic Anemia Study (IAAS) involving several regions in Europe and Israel was published in 1987 and established a well-accepted AA incidence of ~2 cases per million people per year9. Several other studies
from the UK14, France15, Spain16, and Brazil17 have shown similar
results. However, recent studies from Asia show a much higher incidence of up to 7 cases per million people per year,18-22 which are
mostly associated with agricultural substances19,21. It is also well known
that the incidence of AA has two peaks, one between 15–30 years of age and another after 60 years of age15,16. The incidence for men and women
appears to be similar but has been reported to be slightly higher among women in some studies9,14.
1.3 Pathophysiology
Most cases of AA can be pathophysiologically characterized by the T cell-mediated destruction of bone marrow hematopoietic cells23.
However, hematopoietic failure can also result from direct physical or chemical damage to the marrow. Radiation and cytostatic drugs used to treat cancer often cause aplasia in the bone marrow, but other treatments, such as nonsteroidal anti-inflammatory drugs, antibiotics (sulfonamides), antithyroid drugs, antiepileptics, psychotropics, gold, penicillamine, and allopurinol, have also been associated with AA development24,25. Childhood AA may follow viral exposure. For
example, the Epstein-Barr virus and seronegative hepatitis can cause SAA26,27, likely by activating the immune system. In addition, other
immunological conditions, such as eosinophilic fasciitis28, thymoma,29
and Graft-versus-Host disease (GvHD)30, can be complicated with AA.
1.3.1 Immunological mechanisms
The idea that immunological mechanisms are responsible for the development of AA was assumed many years ago (1960–1970s) based on the first bone marrow transplantation results. The conditioning regimen with an anti-lymphocytic serum that enhanced the transplantation results (reduced acute secondary disease)31 and failed
attempts of the syngeneic transplants without conditioning showed that the substitution of bone marrow/stem cells alone could not repopulate the bone marrow32. Even if some immunological mechanisms
contributing to the development of AA have been identified, the responsiveness to immunosuppressive treatment (IST) remains the most convincing evidence of an underlying immune pathophysiology33.
Already in the 1990s, a deficit of early progenitor cells was found, proposing that less than 10% of healthy stem cells are left at the time of the clinical presentation of AA23,34. Moreover, research has focused on
cytotoxic T cells, which appear to be functionally and phenotypically activated35-40, producing interferon-γ (IFN-γ)41,42 and stimulating
1.1 Definition and diagnostic criteria
AA is sometimes referred to as idiopathic, acquired, or immune. It is defined as pancytopenia with hypocellular bone marrow in the absence of fibrosis or an abnormal infiltrate. For the diagnosis of AA, at least two of the three following criteria must be present: hemoglobin <100 g/l, platelet count <50×109/l, or neutrophil count <1.5×109/l8,9.
To assess AA severity, modified Camitta criteria are used8,10:
Severe AA (SAA) is defined as bone marrow cellularity <25% or 25%– 50% with <30% residual hematopoietic cells and two of the following three criteria: neutrophil count <0.5×109/l, platelet count <20×109/l, or
reticulocyte count <20×109/l. Very severe AA (VSAA) is similar to
SAA but with a neutrophil count <0.2×109/l. Non-severe AA (NSAA)
does not fulfill the criteria for SAA or VSAA.
1.2 Epidemiology
The incidence of AA varies considerably in different parts of the world and has also changed over the years. Historically, reports from Europe and the United States in 1960–70 showed diverse results; incidence rates of up to 6–10 AA cases per million people per year. In some of these studies, an association with toxic agriculture substances was shown11-13.
However, different diagnostic criteria were used, and some cases represented other diagnoses11. The International Agranulocytosis and
Aplastic Anemia Study (IAAS) involving several regions in Europe and Israel was published in 1987 and established a well-accepted AA incidence of ~2 cases per million people per year9. Several other studies
from the UK14, France15, Spain16, and Brazil17 have shown similar
results. However, recent studies from Asia show a much higher incidence of up to 7 cases per million people per year,18-22 which are
mostly associated with agricultural substances19,21. It is also well known
that the incidence of AA has two peaks, one between 15–30 years of age and another after 60 years of age15,16. The incidence for men and women
appears to be similar but has been reported to be slightly higher among women in some studies9,14.
1.3 Pathophysiology
Most cases of AA can be pathophysiologically characterized by the T cell-mediated destruction of bone marrow hematopoietic cells23.
However, hematopoietic failure can also result from direct physical or chemical damage to the marrow. Radiation and cytostatic drugs used to treat cancer often cause aplasia in the bone marrow, but other treatments, such as nonsteroidal anti-inflammatory drugs, antibiotics (sulfonamides), antithyroid drugs, antiepileptics, psychotropics, gold, penicillamine, and allopurinol, have also been associated with AA development24,25. Childhood AA may follow viral exposure. For
example, the Epstein-Barr virus and seronegative hepatitis can cause SAA26,27, likely by activating the immune system. In addition, other
immunological conditions, such as eosinophilic fasciitis28, thymoma,29
and Graft-versus-Host disease (GvHD)30, can be complicated with AA.
1.3.1 Immunological mechanisms
The idea that immunological mechanisms are responsible for the development of AA was assumed many years ago (1960–1970s) based on the first bone marrow transplantation results. The conditioning regimen with an anti-lymphocytic serum that enhanced the transplantation results (reduced acute secondary disease)31 and failed
attempts of the syngeneic transplants without conditioning showed that the substitution of bone marrow/stem cells alone could not repopulate the bone marrow32. Even if some immunological mechanisms
contributing to the development of AA have been identified, the responsiveness to immunosuppressive treatment (IST) remains the most convincing evidence of an underlying immune pathophysiology33.
Already in the 1990s, a deficit of early progenitor cells was found, proposing that less than 10% of healthy stem cells are left at the time of the clinical presentation of AA23,34. Moreover, research has focused on
cytotoxic T cells, which appear to be functionally and phenotypically activated35-40, producing interferon-γ (IFN-γ)41,42 and stimulating
shown that macrophages (MΦs) play a central role in IFN-γ-related stem cell destruction. Specifically, they produce tumor necrosis factor-a (TNF-a), engage cytotoxic T cells, and consequently enhance the production of IFN-γ, forming an interactive loop44-46. Another important
aspect involves regulatory T cells (Tregs), which are decreased in AA. There are some indications that the number of Tregs or a unique subtype of Tregs is correlated with the response to ATG treatment, and Tregs appear to increase with the response to treatment47-49. Interleukin
(IL)-17-producing CD4+ cells (Th17) play a pivotal role in autoimmunity50,
and although AA does not share many similarities with autoimmune diseases, Th17 cells are increased (SAA) and seem to play a role during the early stage of AA51,52.
1.3.2 Telomeres
Even though immune destruction of hematopoietic cells is the leading cause of AA, up to 1/3 of patients have short telomeres, and in some cases, these are caused by mutations in the telomerase gene complex53,54.
Telomeres are nucleotide repeats at the ends of chromosomes, which protect them from damage. Telomere maintenance requires the telomerase ribonucleoprotein complex, which consists of telomerase reverse transcriptase (TERT) and its RNA template (TERC)55,56.
Originally, telomere shortening was thought to be the result of a higher than normal number of progenitor cell divisions to produce a mature cell53,57. Nevertheless, discoveries in congenital AAs, such as
dyskeratosis congenita (DC), Fanconi anemia (FA), and Shwachman-Diamond syndrome, helped to connect genetic lesions and telomere shortening of hematopoietic cells58,59. Mutations in TERT and TERC
genes have been found in 4%–10% of AA patients53,60. AA patients with
short telomeres appear to have dismal overall survival (OS) after IST and a significantly higher risk for relapse and clonal evolution54.
1.3.3 Clonality
Although AA is a non-malignant disease, it has a well-accepted association with clonal hematopoiesis61,62. The most used phrase is
clonal evolution, which can occur many years after the initial diagnosis and describes when AA transforms to a myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). In different studies, the long-term risk for developing AML or MDS ranged between 5%–20%, and it often involves the loss of all or a portion of chromosome 763-68.
However, it is not uncommon for AA patients to have chromosomal abnormalities at the time of diagnosis or acquire them during the course of the disease69,70 without further development of malignancy3,68. The
responses to IST and survival appear to be similar between patients with and without cytogenetic aberrations71, but patients with monosomy 7
respond poorly to IST70.
Next-generation sequencing has identified several mutations in specific genes, adding evidence of the clonal nature of AA. Most of the mutations are known from myeloid malignancies, but AA differs from malignancies in its limited set of mutations and clone size72,73. Previous
studies have shown that clonal populations are present in 19% to 47% of AA patients, and the common somatic mutations are DNMT3A,
ASXL1, BCOR, BCORL1, and PIGA72,73. These mutations have also
been associated with the clinical outcome and prognosis; patients with
PIGA, BCOR, and BCORL1 respond well to IST and have better OS
compared with patients with DNMT3A and ASXL1 mutations72. In
another study, the presence of DNMT3A and ASXL1 increased the risk for MDS development73.
Paroxysmal nocturnal hemoglobinuria (PNH) provides historical and considerable evidence of benign clonal hematopoiesis61. PNH is caused
by mutations in the X-linked gene PIGA, which encodes phosphatidylinositol N-acetylglucosaminyltransferase subunit A (PIGA) involved in the synthesis of glycosylphosphatidylinositol (GPI),
shown that macrophages (MΦs) play a central role in IFN-γ-related stem cell destruction. Specifically, they produce tumor necrosis factor-a (TNF-a), engage cytotoxic T cells, and consequently enhance the production of IFN-γ, forming an interactive loop44-46. Another important
aspect involves regulatory T cells (Tregs), which are decreased in AA. There are some indications that the number of Tregs or a unique subtype of Tregs is correlated with the response to ATG treatment, and Tregs appear to increase with the response to treatment47-49. Interleukin
(IL)-17-producing CD4+ cells (Th17) play a pivotal role in autoimmunity50,
and although AA does not share many similarities with autoimmune diseases, Th17 cells are increased (SAA) and seem to play a role during the early stage of AA51,52.
1.3.2 Telomeres
Even though immune destruction of hematopoietic cells is the leading cause of AA, up to 1/3 of patients have short telomeres, and in some cases, these are caused by mutations in the telomerase gene complex53,54.
Telomeres are nucleotide repeats at the ends of chromosomes, which protect them from damage. Telomere maintenance requires the telomerase ribonucleoprotein complex, which consists of telomerase reverse transcriptase (TERT) and its RNA template (TERC)55,56.
Originally, telomere shortening was thought to be the result of a higher than normal number of progenitor cell divisions to produce a mature cell53,57. Nevertheless, discoveries in congenital AAs, such as
dyskeratosis congenita (DC), Fanconi anemia (FA), and Shwachman-Diamond syndrome, helped to connect genetic lesions and telomere shortening of hematopoietic cells58,59. Mutations in TERT and TERC
genes have been found in 4%–10% of AA patients53,60. AA patients with
short telomeres appear to have dismal overall survival (OS) after IST and a significantly higher risk for relapse and clonal evolution54.
1.3.3 Clonality
Although AA is a non-malignant disease, it has a well-accepted association with clonal hematopoiesis61,62. The most used phrase is
clonal evolution, which can occur many years after the initial diagnosis and describes when AA transforms to a myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). In different studies, the long-term risk for developing AML or MDS ranged between 5%–20%, and it often involves the loss of all or a portion of chromosome 763-68.
However, it is not uncommon for AA patients to have chromosomal abnormalities at the time of diagnosis or acquire them during the course of the disease69,70 without further development of malignancy3,68. The
responses to IST and survival appear to be similar between patients with and without cytogenetic aberrations71, but patients with monosomy 7
respond poorly to IST70.
Next-generation sequencing has identified several mutations in specific genes, adding evidence of the clonal nature of AA. Most of the mutations are known from myeloid malignancies, but AA differs from malignancies in its limited set of mutations and clone size72,73. Previous
studies have shown that clonal populations are present in 19% to 47% of AA patients, and the common somatic mutations are DNMT3A,
ASXL1, BCOR, BCORL1, and PIGA72,73. These mutations have also
been associated with the clinical outcome and prognosis; patients with
PIGA, BCOR, and BCORL1 respond well to IST and have better OS
compared with patients with DNMT3A and ASXL1 mutations72. In
another study, the presence of DNMT3A and ASXL1 increased the risk for MDS development73.
Paroxysmal nocturnal hemoglobinuria (PNH) provides historical and considerable evidence of benign clonal hematopoiesis61. PNH is caused
by mutations in the X-linked gene PIGA, which encodes phosphatidylinositol N-acetylglucosaminyltransferase subunit A (PIGA) involved in the synthesis of glycosylphosphatidylinositol (GPI),
making the cell surface extremely sensitive to complement attack74,75.
PNH is strongly associated with bone marrow failure. It is believed that the cell-mediated autoimmune attack that causes AA may spare selectively GPI-negative hematopoietic stem cells, providing a growth advantage for the PIGA- mutant GPI-negative clone74. More than 50%
of AA patients have a PNH clone, most of which have a clone size of <10%, and the development to a symptomatic PNH disease occurs in 2%–19% of AA cases76-80. Nevertheless, the progression to acute
leukemia in patients with PNH and from PNH clones in patients with marrow failure is rare61,81.
AA development has also been associated with specific human leukocyte antigens (HLA)82. Class I HLA molecules, such as HLA
A02:01, A02:06, A31:01, and B4:02, can act as auto-antigens and are targeted by cytotoxic T cells83. A substantial proportion of AA patients
have clonal hematopoiesis characterized by the loss of specific HLA alleles as a result of acquired copy number-neutral loss of heterozygosity (CNN-LOH) on the short arm of chromosome 6 (6pLOH)83. 6pLOH(+)
clones have been detected in 11%–13% of AA patients. Reduced HLA auto-antigen expression gives stem cells an opportunity for clonal outgrowth as an “escape” from the autoimmune attack82-86. This may be
the mechanism by which hematopoiesis is maintained for years in AA patients83.
1.4 Differential diagnosis
AA is a diagnosis of exclusion, and there is no single test that reliably establishes an AA diagnosis (Figure 1). Other causes of pancytopenia
(e.g., drugs or viruses), hypoplastic bone marrow, and inherited bone marrow failure syndromes (IBMFS) should be excluded.
Figure 1. Differential diagnosis of aplastic anemia. Reproduced with
permission from (Young, N; Aplastic anemia; N Engl J Med
2018;379;1643-1656). Copyright Massachusetts Medical Society.
1.4.1 Inherited bone marrow failure syndromes
IBMFS are traditionally considered pediatric disorders, but they can also be diagnosed in adults. IBMFS are rare disorders, but the most common ones to develop to AA and evolve into MDS and AML are FA, DC, and
GATA2 spectrum disorders87. Patients with IBMFS often have both
physical and hematologic findings, but depending on the disease, they may not have any physical abnormalities.
making the cell surface extremely sensitive to complement attack74,75.
PNH is strongly associated with bone marrow failure. It is believed that the cell-mediated autoimmune attack that causes AA may spare selectively GPI-negative hematopoietic stem cells, providing a growth advantage for the PIGA- mutant GPI-negative clone74. More than 50%
of AA patients have a PNH clone, most of which have a clone size of <10%, and the development to a symptomatic PNH disease occurs in 2%–19% of AA cases76-80. Nevertheless, the progression to acute
leukemia in patients with PNH and from PNH clones in patients with marrow failure is rare61,81.
AA development has also been associated with specific human leukocyte antigens (HLA)82. Class I HLA molecules, such as HLA
A02:01, A02:06, A31:01, and B4:02, can act as auto-antigens and are targeted by cytotoxic T cells83. A substantial proportion of AA patients
have clonal hematopoiesis characterized by the loss of specific HLA alleles as a result of acquired copy number-neutral loss of heterozygosity (CNN-LOH) on the short arm of chromosome 6 (6pLOH)83. 6pLOH(+)
clones have been detected in 11%–13% of AA patients. Reduced HLA auto-antigen expression gives stem cells an opportunity for clonal outgrowth as an “escape” from the autoimmune attack82-86. This may be
the mechanism by which hematopoiesis is maintained for years in AA patients83.
1.4 Differential diagnosis
AA is a diagnosis of exclusion, and there is no single test that reliably establishes an AA diagnosis (Figure 1). Other causes of pancytopenia
(e.g., drugs or viruses), hypoplastic bone marrow, and inherited bone marrow failure syndromes (IBMFS) should be excluded.
Figure 1. Differential diagnosis of aplastic anemia. Reproduced with
permission from (Young, N; Aplastic anemia; N Engl J Med
2018;379;1643-1656). Copyright Massachusetts Medical Society.
1.4.1 Inherited bone marrow failure syndromes
IBMFS are traditionally considered pediatric disorders, but they can also be diagnosed in adults. IBMFS are rare disorders, but the most common ones to develop to AA and evolve into MDS and AML are FA, DC, and
GATA2 spectrum disorders87. Patients with IBMFS often have both
physical and hematologic findings, but depending on the disease, they may not have any physical abnormalities.
Fanconi anemia
FA is the most common inherited cause of bone marrow failure. Genes that are mutated in FA patients are termed FANC, and more than 20 have been identified; FANCA, FANCC, FANCG, and FANCD2 are the most common88,89. FA patients have a cumulative hematopoietic dysfunction
likely related to an excess of genetic instability, cellular stress, p53 activation, and cell death90. Patients often present with congenital
abnormalities, such as a short statue, microphthalmia, thumb and radius deformities, skin hyperpigmentation, and other signs, but approximately 25% have no physical signs, and the disease can be missed during childhood. Up to 80% of FA patients have been reported to develop bone marrow failure, and it usually occurs during the first or second decades of life 90,91. MDS or AML is diagnosed later on in the disease course,
and the risk to develop solid tumors increases with age92.
Dyskeratosis congenita
DC is a disease in which many patients reach adulthood before diagnosis. Possible physical signs increase with age, and more than half of the patients are diagnosed later than 15 years of age87. Dysplastic
nails, oral leukoplakia, hypogonadism, lacrimal duct stenosis, early greying, and liver and pulmonary fibrosis can all be symptoms of DC. Since 1998, eleven DC genes have been identified89. Several of them are
essential in telomere maintenance, which becomes defective and leads to very short telomeres. DC can have different inheritance patterns.
DKC1 is responsible for X-linked DC and often found in children89.
Autosomal dominant DC is heterogeneous, and three mutations, including TERC, TERT, and TINF2, have been recognized89.
Autosomal-recessive DC has been associated with NOP10 and NHP2 mutations93. Bone marrow failure is a frequent complication in DC (in
up to 80% of patients), and it often develops in the second or third decade but can develop from any time after birth to the seventh decade of life87,93. DC patients generally have a 4-fold higher risk for
developing malignancies, but a substantially higher risk is noted for
some types of tumors, including head and neck carcinomas (~70-fold), MDS (~500-fold), and AML (~70-fold)94.
GATA2 spectrum disorders
GATA2 is a transcription factor essential for hematopoiesis by maintaining the pool of hematopoietic stem cells95. GATA2 deficiency
has a broad clinical spectrum that includes viral and bacterial infections, lymphedema, deafness, alveolar proteinosis, monocytopenia, and low B, T, and NK cells, and it is associated with bone marrow failure, MDS, and AML96. GATA2 mutations associated with bone marrow failure
result in loss of function and GATA2 haploinsufficiency89. Patients with
GATA2 deficiency have a high risk of developing MDS and leukemia97,
and MDS is typically hypocellular. Additionally, the most common cytogenetic aberration is monosomy 7, but other abnormalities, such as trisomy 8 and trisomy 21, have also been described95. Flow cytometry
analysis of GATA2 patient bone marrow samples showed a concomitant reduction in monocytes, B cells, and NK cells, which were significantly lower than in AA patients98.
1.4.2 Hypoplastic myelodysplastic syndrome
The diagnosis of a MDS requires morphological dysplasia in 10% of one or more myeloid lineages99, and most patients have a hypercellular
or normocellular marrow. Still, 10%–20% of MDS patients have cellularity under 30%, having a hypoplastic MDS (hMDS)100,101.
Distinguishing hMDS from AA can be difficult; using only morphological alterations is challenging because of the low number of assessable cells. The detection of clonal markers as a chromosomal aberration or molecular marker can be helpful, but some overlapping with AA occurs102. However, dysplastic megakaryocytes, a normal or
increased number of CD34 cells, and some mutations are not consistent with AA103-106. Spliceosome genes mutations and multiple mutated
Fanconi anemia
FA is the most common inherited cause of bone marrow failure. Genes that are mutated in FA patients are termed FANC, and more than 20 have been identified; FANCA, FANCC, FANCG, and FANCD2 are the most common88,89. FA patients have a cumulative hematopoietic dysfunction
likely related to an excess of genetic instability, cellular stress, p53 activation, and cell death90. Patients often present with congenital
abnormalities, such as a short statue, microphthalmia, thumb and radius deformities, skin hyperpigmentation, and other signs, but approximately 25% have no physical signs, and the disease can be missed during childhood. Up to 80% of FA patients have been reported to develop bone marrow failure, and it usually occurs during the first or second decades of life 90,91. MDS or AML is diagnosed later on in the disease course,
and the risk to develop solid tumors increases with age92.
Dyskeratosis congenita
DC is a disease in which many patients reach adulthood before diagnosis. Possible physical signs increase with age, and more than half of the patients are diagnosed later than 15 years of age87. Dysplastic
nails, oral leukoplakia, hypogonadism, lacrimal duct stenosis, early greying, and liver and pulmonary fibrosis can all be symptoms of DC. Since 1998, eleven DC genes have been identified89. Several of them are
essential in telomere maintenance, which becomes defective and leads to very short telomeres. DC can have different inheritance patterns.
DKC1 is responsible for X-linked DC and often found in children89.
Autosomal dominant DC is heterogeneous, and three mutations, including TERC, TERT, and TINF2, have been recognized89.
Autosomal-recessive DC has been associated with NOP10 and NHP2 mutations93. Bone marrow failure is a frequent complication in DC (in
up to 80% of patients), and it often develops in the second or third decade but can develop from any time after birth to the seventh decade of life87,93. DC patients generally have a 4-fold higher risk for
developing malignancies, but a substantially higher risk is noted for
some types of tumors, including head and neck carcinomas (~70-fold), MDS (~500-fold), and AML (~70-fold)94.
GATA2 spectrum disorders
GATA2 is a transcription factor essential for hematopoiesis by maintaining the pool of hematopoietic stem cells95. GATA2 deficiency
has a broad clinical spectrum that includes viral and bacterial infections, lymphedema, deafness, alveolar proteinosis, monocytopenia, and low B, T, and NK cells, and it is associated with bone marrow failure, MDS, and AML96. GATA2 mutations associated with bone marrow failure
result in loss of function and GATA2 haploinsufficiency89. Patients with
GATA2 deficiency have a high risk of developing MDS and leukemia97,
and MDS is typically hypocellular. Additionally, the most common cytogenetic aberration is monosomy 7, but other abnormalities, such as trisomy 8 and trisomy 21, have also been described95. Flow cytometry
analysis of GATA2 patient bone marrow samples showed a concomitant reduction in monocytes, B cells, and NK cells, which were significantly lower than in AA patients98.
1.4.2 Hypoplastic myelodysplastic syndrome
The diagnosis of a MDS requires morphological dysplasia in 10% of one or more myeloid lineages99, and most patients have a hypercellular
or normocellular marrow. Still, 10%–20% of MDS patients have cellularity under 30%, having a hypoplastic MDS (hMDS)100,101.
Distinguishing hMDS from AA can be difficult; using only morphological alterations is challenging because of the low number of assessable cells. The detection of clonal markers as a chromosomal aberration or molecular marker can be helpful, but some overlapping with AA occurs102. However, dysplastic megakaryocytes, a normal or
increased number of CD34 cells, and some mutations are not consistent with AA103-106. Spliceosome genes mutations and multiple mutated
have a higher risk of developing leukemia and shorter OS compared with AA106,108.
Differentiating between pediatric bone marrow failure syndromes is challenging. The most common form of pediatric MDS is refractory cytopenia characterized by a hypocellular bone marrow with a normal karyotype109,110. Somatic mutations and inherited constitutional
mutations should be analyzed for correct diagnosis and treatment89,110.
1.5 Treatment
Patients with SAA or VSAA have a direct indication of treatment. NSAA has a different clinical course of the disease as they can develop SAA or be stable for years. Spontaneous recoveries occur but are rare111,112. NSAA patients can be monitored with blood tests and treated
when they become dependent on transfusions or have a neutrophil count <0.5×109/l113. Current treatment recommendations stratify patients
according to age and the availability of an HLA-matched sibling donor
(Figure 2)3,113,114. However, the choice between hematopoietic stem cell
transplantation (HSCT) and IST is not always clear-cut, and both comorbidities and the severity of AA should be taken into account115.
Figure 2. Treatment of acquired aplastic anemia; Reprinted from, Brit. J.
Haematol;172(2), Killick, B.S, et al; Guidelines for the diagnosis and management of adult aplastic anaemia, Pages No 187-207, Copyright (2016), with permission from Elsevier.
have a higher risk of developing leukemia and shorter OS compared with AA106,108.
Differentiating between pediatric bone marrow failure syndromes is challenging. The most common form of pediatric MDS is refractory cytopenia characterized by a hypocellular bone marrow with a normal karyotype109,110. Somatic mutations and inherited constitutional
mutations should be analyzed for correct diagnosis and treatment89,110.
1.5 Treatment
Patients with SAA or VSAA have a direct indication of treatment. NSAA has a different clinical course of the disease as they can develop SAA or be stable for years. Spontaneous recoveries occur but are rare111,112. NSAA patients can be monitored with blood tests and treated
when they become dependent on transfusions or have a neutrophil count <0.5×109/l113. Current treatment recommendations stratify patients
according to age and the availability of an HLA-matched sibling donor
(Figure 2)3,113,114. However, the choice between hematopoietic stem cell
transplantation (HSCT) and IST is not always clear-cut, and both comorbidities and the severity of AA should be taken into account115.
Figure 2. Treatment of acquired aplastic anemia; Reprinted from, Brit. J.
Haematol;172(2), Killick, B.S, et al; Guidelines for the diagnosis and management of adult aplastic anaemia, Pages No 187-207, Copyright (2016), with permission from Elsevier.
1.5.1 Hematopoietic stem cell transplantation
The replacement of failing bone marrow with a healthy one is a curative treatment for AA patients. HSCT has several complications, and GvHD remains one of the most feared reactions. It is the major cause of non-relapse mortality after HSCT. Patients with severe chronic (c)GvHD have decreased quality of life, and the long-term mortality rate increases to 50%116,117.
The choice of primary treatment depends on the availability of an HLA-matched sibling donor, and patients <40 (-50) years of age with a sibling donor should undergo transplantation as the first-line treatment113.
These recommendations have been changed over the years. As early as 1988, it was established that patients <20 years of age had better OS with an HLA-matched sibling donor HSCT than IST (66% vs. 56%); this was especially true for patients with VSAA (64% vs. 38%, p=0.01). Older patients with VSAA responded poorly to both treatments (OS 44% HSCT vs. 43% IST), but SAA patients had better survival with IST (82%) than HSCT (62%)10. In 1995, HSCT was also suggested for
patients 20–45 years of age27. However, even though studies still show
exceptionally promising HSCT results for younger patients, patients >40 years of age have a considerably inferior outcome. A European Group for Blood and Marrow Transplantation (EBMT) study from the year 2000 showed better results for young (<20 years of age) patients with low neutrophil counts (<0.2×109/l) receiving HSCT118. A follow-up
study from 2016 that compared the findings from before and after 1999 confirmed the superior results for the younger group, in which the OS after HSCT was 86%, and patients 21–40 years of age also had a high OS with HSCT (76% compared with 65% for IST, p=0.6). Nevertheless, patients >40 years of age showed a worse OS with HSCT (IST 58% and HSCT 56%, p=0.006)119. Patients >40 years of age had a declining OS:
40–49 years 67%, 50–59 years 58%, and >60 years 48% (p<0.0001)120.
It is well-known that older patients have higher transplant-related mortality (TRM), and a low performance score pre-transplant has been
shown to increase the mortality risk121. As infections, GvHD, and
toxicity are the main reasons for death, questions of the most appropriate conditioning regimen and the optimal GvHD prophylaxis have been raised120,122,123.
Bone marrow (BM) has been widely used as a stem cell source for HSCT in AA patients. At the beginning of the 21st century, when peripheral blood stem cells (PBSCs) were being increasingly used, a switch from BM to PBSC occurred. This also led to studies comparing the transplant results using BM or PBSCs. A large study combining data from EBMT and the Center for International Blood and Marrow Transplant Research (CIBMTR) was published in 2007 and included 692 AA patients. A clear advantage for BM was observed in the <20 years of age group, with a better OS (85% vs. 73%), as well as a lower risk for cGvHD124. In another study, the survival of patients receiving
BM was 84% compared with 68% for PBSCs (p<0.0001), and this advantage was present in all age groups (<20 years: 90% vs. 76% and >20 years: 74% vs. 64%, respectively)125. As a result, BM is currently
the preferred stem cell source for transplantation in AA113, even though
some studies reported no differences in OS or GvHD using PBSCs instead of BM120.
As only a few patients have an HLA-matched sibling donor, the first attempts to use unrelated donors (URDs) was made in the 1980s. The probability of long-term survival after an URD transplantation in AA ranged from 29% to 50%126-129. Since then, the results have improved
with the introduction of high-resolution HLA typing and the use of fludarabine-based conditioning, low dose total body irradiation, and the CD52-antibody alemtuzumab130-133. An EBMT study on SAA patients
transplanted between 1990 and 2005 found that cohorts transplanted before and after 1998 significantly differed in outcomes. The 5-year survival increased from 32% before 1998 to 57% for those transplanted after (p<0.0001), and the improved survival was associated with less graft failure and reduced acute and chronic GvHD134. Another EBMT
1.5.1 Hematopoietic stem cell transplantation
The replacement of failing bone marrow with a healthy one is a curative treatment for AA patients. HSCT has several complications, and GvHD remains one of the most feared reactions. It is the major cause of non-relapse mortality after HSCT. Patients with severe chronic (c)GvHD have decreased quality of life, and the long-term mortality rate increases to 50%116,117.
The choice of primary treatment depends on the availability of an HLA-matched sibling donor, and patients <40 (-50) years of age with a sibling donor should undergo transplantation as the first-line treatment113.
These recommendations have been changed over the years. As early as 1988, it was established that patients <20 years of age had better OS with an HLA-matched sibling donor HSCT than IST (66% vs. 56%); this was especially true for patients with VSAA (64% vs. 38%, p=0.01). Older patients with VSAA responded poorly to both treatments (OS 44% HSCT vs. 43% IST), but SAA patients had better survival with IST (82%) than HSCT (62%)10. In 1995, HSCT was also suggested for
patients 20–45 years of age27. However, even though studies still show
exceptionally promising HSCT results for younger patients, patients >40 years of age have a considerably inferior outcome. A European Group for Blood and Marrow Transplantation (EBMT) study from the year 2000 showed better results for young (<20 years of age) patients with low neutrophil counts (<0.2×109/l) receiving HSCT118. A follow-up
study from 2016 that compared the findings from before and after 1999 confirmed the superior results for the younger group, in which the OS after HSCT was 86%, and patients 21–40 years of age also had a high OS with HSCT (76% compared with 65% for IST, p=0.6). Nevertheless, patients >40 years of age showed a worse OS with HSCT (IST 58% and HSCT 56%, p=0.006)119. Patients >40 years of age had a declining OS:
40–49 years 67%, 50–59 years 58%, and >60 years 48% (p<0.0001)120.
It is well-known that older patients have higher transplant-related mortality (TRM), and a low performance score pre-transplant has been
shown to increase the mortality risk121. As infections, GvHD, and
toxicity are the main reasons for death, questions of the most appropriate conditioning regimen and the optimal GvHD prophylaxis have been raised120,122,123.
Bone marrow (BM) has been widely used as a stem cell source for HSCT in AA patients. At the beginning of the 21st century, when peripheral blood stem cells (PBSCs) were being increasingly used, a switch from BM to PBSC occurred. This also led to studies comparing the transplant results using BM or PBSCs. A large study combining data from EBMT and the Center for International Blood and Marrow Transplant Research (CIBMTR) was published in 2007 and included 692 AA patients. A clear advantage for BM was observed in the <20 years of age group, with a better OS (85% vs. 73%), as well as a lower risk for cGvHD124. In another study, the survival of patients receiving
BM was 84% compared with 68% for PBSCs (p<0.0001), and this advantage was present in all age groups (<20 years: 90% vs. 76% and >20 years: 74% vs. 64%, respectively)125. As a result, BM is currently
the preferred stem cell source for transplantation in AA113, even though
some studies reported no differences in OS or GvHD using PBSCs instead of BM120.
As only a few patients have an HLA-matched sibling donor, the first attempts to use unrelated donors (URDs) was made in the 1980s. The probability of long-term survival after an URD transplantation in AA ranged from 29% to 50%126-129. Since then, the results have improved
with the introduction of high-resolution HLA typing and the use of fludarabine-based conditioning, low dose total body irradiation, and the CD52-antibody alemtuzumab130-133. An EBMT study on SAA patients
transplanted between 1990 and 2005 found that cohorts transplanted before and after 1998 significantly differed in outcomes. The 5-year survival increased from 32% before 1998 to 57% for those transplanted after (p<0.0001), and the improved survival was associated with less graft failure and reduced acute and chronic GvHD134. Another EBMT
analysis of 1448 AA patients transplanted between 2005 and 2009 included 508 URD transplants, which showed significantly more acute grade II–IV (25% vs 14%) and chronic GvHD (26% vs 14%). In survival analyses, URD compared with an HLA-matched sibling donor graft was not a statistically significant predictor135. The OS after an URD graft is
currently >70% for adult patients (up to 40 years of age) and >90% for children (<18 years of age)136-139. Therefore, URD transplantation is
recommended as a second-line treatment for adults but can be used as a first-line treatment for children137.
In addition to well-defined transplantation options with an HLA-matched sibling donor or an URD, alternative stem cell sources, such as cord blood or a haploidentical donor, have been used. These may also be curative, but the risk of graft rejection, infectious complications, and GvHD are higher140. A large cohort study of unrelated cord blood
transplantation between 1996 and 2009 showed an estimated 3-year OS of 38%. The major cause of death was graft failure and infections141.
More recently, an increasingly used option is haploidentical transplantation. Almost 400 AA patients have been transplanted, and encouraging results have been reported mainly in the pediatric population and young adults142. The median overall engraftment rate
was 92%, and the median risk of grade 2–4 acute GvHD was 12%, with a median 1-year OS of 85%143.
Apart from developing acute or chronic GvHD, HSCT is associated with other potential complications. Rejections are observed in 5%–15% of AA patients and can be reduced with the use of ATG, low-dose irradiation, fludarabine-based conditioning for URD grafts, and using a marrow cell dose higher than 2×108/kg114. Late complications include
the development of solid tumors, the incidence of which has been reported to be 12%, and the most prominent locations were the skin, cervix, head and neck. The conditioning regimen, especially irradiation, and cGvHD were risk factors. Another late complication is
osteonecrosis, with an incidence of 15%–20%, and the risk factors were age, previous IST, and treatment with steroids144-146.
1.5.2 Immunosuppressive treatment
Today, the standard IST is horse anti-thymocyte globulin (hATG) and cyclosporine A (CsA), with hematologic recovery in 50% to 70% of cases77,114,147-151. Adding CsA to hATG considerably improved the
response rate at three months (65% in the hATG/CsA group vs. 39% in the hATG group) and six months (70% vs. 46%, respectively)147.
However, later attempts to enhance hATG/CsA treatment by adding an additional immunosuppressive drug have been unsuccessful. Mycophenolate mofetil (MMF) inhibits the proliferation of activated lymphocytes and was anticipated to favor the induction of tolerance. The OR rate at six months was 62% with a 37% relapse rate, and more than half of the relapses occurred during MMF administration, suggesting that this drug was not effective in preventing relapses among responders152. Sirolimus blocks CsA-resistant pathways and inhibits T
cell activation. A study in which sirolimus was added to hATG + CsA showed no differences in the OR rate at either three or six months (37% and 51% for the hATG/CsA/sirolimus group and 50% and 62% for the hATG/CsA group, respectively)153. Alternative treatment regimens,
such as alemtuzumab and cyclophosphamide, have also been used for AA treatment. As first-line treatment, alemtuzumab achieved only a 19% response rate154 and is not currently recommended.
Cyclophosphamide administered at moderate or high (30–200 mg/kg) doses is reported to be too toxic, causing prolonged neutropenia and a high incidence of fungal infections and mortality155,156.
IST is offered as a first-line treatment to patients without an HLA-matched sibling donor or those >40 years of age113. There are several
ATG products available with different efficacies. All ATG products have a lymphocyte-depleting effect, but rabbit ATG (rATG) causes more severe depletion of CD4+ T cells and Tregs157. However, this
